Application Security

AI Poised to Drive New Wave of Exploits

Criminals are ready to use AI to dramatically speed the process of finding zero-day vulnerabilities in systems.

Artificial intelligence has significant potential for use in cybersecurity – on both sides of the security battle lines. And you don't have to wait for scenarios out of "The Terminator" to see its impact.

According to Derek Manky, Fortinet's chief, security insights & global threat alliances, AI's use by attackers is a simple matter of economics. "Looking forward, cybercriminals will be looking at increasing their ROI. I think what we'll start to see is the concept of AI fuzzing," he says. "We've seen some interesting research on this."

"Fuzzing" – among a series of predictions in Fortinet's Q3 2018 "Security Prediction Report" – is a technique that has its roots in software quality testing.   The system (or component) being tested is given random input until it crashes, and then the crash is analyzed. From an attacker's point of view, fuzzing can uncover vulnerabilities to exploit.

"Discovering vulnerabilities is still quite manual, very human-driven," Manky says, which makes finding them an expensive process. AI can dramatically speed up the fuzzing process, and it's already been proved in other contexts. "Groups have applied this to gaming, using AI to hack the game and exploit a weakness," Manky says. Moving that experience to finding security exploits is a natural next step.

Attackers can use AI to dramatically shorten the time from finding a problem to creating an exploit, as well. Groups will be "using AI to study code and systems to find vulnerabilities, and then using AI to find the best exploit of those vulnerabilities," Manky says. "This is automatically creating zero-days."

AI, in this context, is a tool for finding the vulnerabilities and exploits, not orchestrating attacks. As that tool is used by more criminal organizations, Manky sees detecting and exploiting zero-days becoming faster and easier, with the cost of those exploits becoming lower and lower on the black market. "From a cybercriminal perspective, the AI becomes a commodity with zero-day mining systems. Zero-days become less expensive and more accessible to hackers," he explains.

Defending a more porous and exploited attack surface is a matter of getting all the basics right, Manky says. "From the CISO perspective, it becomes more important for your patch management to be good, and open collaboration becomes more important," he says. In addition, designing zero-trust architectures that are thoroughly segmented is critical. "You don't want a successful attack gaining access to the rest of the network," Manky explains.

Related Content:

 

Black Hat Europe returns to London Dec 3-6 2018  with hands-on technical Trainings, cutting-edge Briefings, Arsenal open-source tool demonstrations, top-tier security solutions and service providers in the Business Hall. Click for information on the conference and to register.

Curtis Franklin Jr. is Senior Editor at Dark Reading. In this role he focuses on product and technology coverage for the publication. In addition he works on audio and video programming for Dark Reading and contributes to activities at Interop ITX, Black Hat, INsecurity, and ... View Full Bio

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Crowdsourced vs. Traditional Pen Testing
Alex Haynes, Chief Information Security Officer, CDL,  3/19/2019
BEC Scammer Pleads Guilty
Dark Reading Staff 3/20/2019
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
Write a Caption, Win a Starbucks Card! Click Here
Latest Comment: This comment is waiting for review by our moderators.
Current Issue
5 Emerging Cyber Threats to Watch for in 2019
Online attackers are constantly developing new, innovative ways to break into the enterprise. This Dark Reading Tech Digest gives an in-depth look at five emerging attack trends and exploits your security team should look out for, along with helpful recommendations on how you can prevent your organization from falling victim.
Flash Poll
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2018-18913
PUBLISHED: 2019-03-21
Opera before 57.0.3098.106 is vulnerable to a DLL Search Order hijacking attack where an attacker can send a ZIP archive composed of an HTML page along with a malicious DLL to the target. Once the document is opened, it may allow the attacker to take full control of the system from any location with...
CVE-2018-20031
PUBLISHED: 2019-03-21
A Denial of Service vulnerability related to preemptive item deletion in lmgrd and vendor daemon components of FlexNet Publisher version 11.16.1.0 and earlier allows a remote attacker to send a combination of messages to lmgrd or the vendor daemon, causing the heartbeat between lmgrd and the vendor ...
CVE-2018-20032
PUBLISHED: 2019-03-21
A Denial of Service vulnerability related to message decoding in lmgrd and vendor daemon components of FlexNet Publisher version 11.16.1.0 and earlier allows a remote attacker to send a combination of messages to lmgrd or the vendor daemon, causing the heartbeat between lmgrd and the vendor daemon t...
CVE-2018-20034
PUBLISHED: 2019-03-21
A Denial of Service vulnerability related to adding an item to a list in lmgrd and vendor daemon components of FlexNet Publisher version 11.16.1.0 and earlier allows a remote attacker to send a combination of messages to lmgrd or the vendor daemon, causing the heartbeat between lmgrd and the vendor ...
CVE-2019-3855
PUBLISHED: 2019-03-21
An integer overflow flaw which could lead to an out of bounds write was discovered in libssh2 before 1.8.1 in the way packets are read from the server. A remote attacker who compromises a SSH server may be able to execute code on the client system when a user connects to the server.